Adds `ALTER TABLE` to the simulator. Currently, there are no properties
that generate `ALTER TABLE`. The query is only generated in
`Property::Query` or in extension queries.
Conditions to generate `ALTER TABLE`:
- In differential testing, do not generate `ALTER COLUMN` as SQLite does
not support it.
- If there is only 1 column, or all columns are present in indexes, do
not generate a `DROP COLUMN` as it would be an error in the database
- if there are no tables, obviously do not generate `ALTER TABLE`
Some fixes:
- handle NULL generation in `GTValue` and `LTValue`, as we now have to
handle nulls due to `ADD COLUMN` adding cols with NULL
- correctly compare NULLs in `binary_compare`
Closes#3650
We currently return the exact same error from two different IdxDelete
paths. Improve the messages with context about what we're doing to make
this error more debuggable.
Reviewed-by: Jussi Saurio <jussi.saurio@gmail.com>
Closes#3699
Add `--busy-timeout` command-line option to turso-stress with a default
value of 5000 ms. This helps prevent spurious database busy errors
during concurrent stress testing and ensure that integrity checks are
not skipped because of concurrent writes.
Closes#3696
We currently return the exact same error from two different IdxDelete
paths. Improve the messages with context about what we're doing to make
this error more debuggable.
Add `--busy-timeout` command-line option to turso-stress with a default
value of 5000 ms. This helps prevent spurious database busy errors
during concurrent stress testing and ensure that integrity checks are
not skipped because of concurrent writes.
I've added a trait called `CursorTrait`. I know it's not a good name for
now, but I didn't know what tto change then enum `Cursor` to. This trait
wraps all common functionality, and some functionality that is yet too
specific that needs to be fixed.
This is needed in order to have layered cursors where for example,
MvccCursor will need a fallback BTreeCursor.
Closes#3660
Prevents something like `WHERE x = 5 AND x = 5` from becoming a two
component index key.
Closes#3656
Reviewed-by: Nikita Sivukhin (@sivukhin)
Closes#3658
## Purpose
- Deploy `tech.turso:turso:<version>` to maven central so that users can
easily use java bindings
- For example :
https://repo1.maven.org/maven2/io/github/seonwkim/turso/0.0.1/
## Requirements
- [x] Add the following github secrets.
- [x] MAVEN_CENTRAL_USERNAME
- [x] MAVEN_CENTRAL_PASSWORD
- [x] GPG_PRIVATE_KEY
- [x] GPG_PASSPHRASE
- [ ] Namespace `tech.turso` must be registered at maven central
- [ ] GPG key registration to key servers
- Notes
- Retrieve MAVEN_CENTRAL_USERNAME and MAVEN_CENTRAL_PASSWORD from
[maven central](https://central.sonatype.com/usertoken)
- GPG keys should be registered. You should distribute your keys to
designated(maven central supported) servers
- Refer to [GPG key related docs](https://central.sonatype.org/publ
ish/requirements/gpg/#distributing-your-public-key)
- Btw, I used `keyserver.ubuntu.com` key server while testing
### [Maven Central Username &
Password](https://central.sonatype.com/usertoken)
<img width="2878" height="1338" alt="image"
src="https://github.com/user-
attachments/assets/03e6f967-a7f6-46b8-aef5-d15772bd9eea" />
### [Maven Central
Namespace](https://central.sonatype.com/publishing/namespaces)
<img width="1424" height="456" alt="image" src="https://github.com/user-
attachments/assets/8c0f4f17-bf5a-4c6a-bc47-748d86cd1f1a" />
## Future Works
- Currently, we depend on gradle.properties to determine the version of
our dependency and it's cumbersome to always change the version
manually. Let's find a better solution.
Closes#3624
This PR introduces sparse vectors support and jaccard distance
implementation.
Also, this PR restructure the code to have all vector operations in
separate files (they grow pretty quickly as new vector representations
added to the DB).
Closes#3647
Various little fixes to `Sorter` that reduce unnecessary work.
Makes TPC-H query 1 roughly 2x faster, which is a lot because it
originally took 30-40 seconds depending on the CI run
Closes#3645
I found an application in the open that expects sqlite_version() to
return a specific string (higher than 3.8...).
We had tons of those issues at Scylla, and the lesson was that you tell
your kids not to lie, but when life hits, well... you lie.
We'll add a new function, turso_version, that tells the truth.
Closes#3635
